Coordinated Transcriptional Regulation of Storage Product ... · PDF fileCoordinated Transcriptional Regulation of Storage Product ... subject of this report. ... formamide and fractionated

$Page 1: Coordinated Transcriptional Regulation of Storage Product ... · PDF fileCoordinated Transcriptional Regulation of Storage Product ... subject of this report. ... formamide and fractionated$
Plant Physiol. (1994) 106: 713-722

Coordinated Transcriptional Regulation of Storage Product Genes in the Maize Endosperm'

Michael J. Ciroux, Charles Boyer, Cunter Feix, and 1. Curtis Hannah* Department of Horticultural Sciences and Program in Plant Molecular and Cellular Biology,

University of Florida, Gainesville, Florida 3261 1 (M.J.G., L.C.H.); Department of Horticulture, Pennsylvania State University, University Park, Pennsylvania 16802 (C.B.); and lnstitut fur Biologie Ill,

Universitat Freiburg, 791 04 Freiburg, Germany (G.F.)

We have demonstrated that expression of genes involved in starch and storage protein synthesis of the maize (Zea mays 1.) endosperm are coordinated. Genetic lesions altering synthetic events in one biosynthetic pathway affect expression of genes in both pathways. Initial studies focused on shrunken2 (sh2) and briffle2 (bf2) mutants because these genes encode subunits of the same enzyme, ADP-glucose pyrophosphorylase. Analysis of various sh2- and bf2- mutant alleles showed that the most severe mutations also conditioned the largest increase in transcripts. The analysis was extended by monitoring the transcripts of the genes, shrunken1 (shl, structural gene for Suc synthase), sh2, bf2, waxy1 (wxl , structural gene for starch synthase), and those of the large and small zeins in isogenic maize lines at 14, 22, and 30 d postpolli- nation. Endosperms were wild type for all of these genes or contained shl-, sh2-, btl-, bt2-, opaque2 (02-), or amylose-extender7 (del-) dulll (dul-) wxl- mutations. Transcripts increased contin- ually throughout kernel development in the mutants relative to the standard W64A used. Variation in the amount of SUC entering the developing seed also altered transcript amounts. The results indi- cate that starch and protein biosynthetic genes act in a concerted manner, and both are sensitive to mutationally induced differences.

The maize (Zea mays L.) endosperm is a specialized tissue that provides a source of nutrients to the germinated embryo. The vast majority of carbon used in the early steps of seedling development is derived from starch and the storage proteins, the zeins. These storage products account for approximately 90% of the dry weight of a mature maize endosperm.

Historically, mutants have provided the most insight into the pathways of starch and zein synthesis. Many endosperm mutations can be classified as affecting starch or zein biosynthesis. Mutations such as 02- , flou y2-, and 07- decrease zein content, and proteins in the other solubility classes increase in abundance (Mertz et al., 1964; Nelson et al., 1965; Ma and Nelson, 1975; Habben et al., 1993). 02 has been shown to encode a trans-acting regulatory protein involved in zein transcription (Schmidt et al., 1990). S and L zeins represent the 19- and 22-kD classes of zein polypeptides and an 02- mutation affects mainly the production of 22-kD zein proteins (Langridge et al., 1982).

Contribution from the Florida Agricultural Experiment Station Journal Series, No. R-03868.

* Corresponding author; fax 1-904-392-6479. 71 3

The majority of the enzymes known to be involved in the starch biosynthetic pathway follow similar developmental timing (Tsai et al., 1970). Ingle et al. (1965) found maximum endosperm sugar content at 22 dpp. This stage of development is also when the starch biosynthetic enzymes are most active (Tsai et al., 1970). Mutations in the genes shl (the structural gene for SUC synthase; Chourey and Nelson, 1976), btl (the structural gene for a membrane-bound, adenylate camer; Sullivan et al., 1991), and sh2 and bt2 (both structural genes for AGP; Hannah and Nelson, 1976) decrease starch content in the mature kemel as a consequence of the loss of an important starch synthetic enzyme or protein. Mutations in wxl (the structural gene for the starch-bound ADP-Glc glucosyl transferase; Nelson and Rines, 1962), ael (the structural gene for a starch-branching enzyme; Boyer and Preiss, 1978), and dul (Creech, 1965) do not reduce starch amount but rather alter the ratio of the two major starch constituents, amylose and amylopectin. The straight-chained amylose is abolished in w x l - mutations, whereas a e l - and d u l - mutations increase the amount of amylose relative to the amount of amylopectin. Neither a e l - nor d u l - reduces the total amount of starch by more than 20% (Creech, 1965). The triple mutant a e l - d u l - wxl- results in kemels with increased sugars and reduced starch similar to sh2-, bt2-, and btl- mutants (Creech, 1965). More detailed descriptions of these mutants can be found in a recent review (Hannah et al., 1993).

Although mutations are historically placed into the starch or zein biosynthetic pathway, hints of possible interactions between these pathways have been noted. For example, Tsai et al. (1978) and Barbosa and Glover (1978) reported that the sh2-, bt2-, and b t l - mutations that severely reduce starch synthesis also reduce zein synthesis to the same extent as 02-.

Ample evidence exists that expression of various genes is modulated by sugar concentration. For example, studies with maize protoplasts revealed repression of transcription of pho- tosynthetic genes by various carbon sources (Sheen, 1990).

Abbreviations: ael, amylose-extenderl; AGP, ADP-Glc pyrophosphorylase; agpl, ADP-Glc pyrophosphorylasel; agp2, ADP-Glc pyrophosphorylase2; b t l , brittlel; bt 2, brittle2; dpp, days postpolli- nation; dul , dulll; L zein, large zein; 02, opaque2; S zein, small zein; shl , shrunkenl; sh2, shrunken2; susl, SUC synthasel; sul, sugaryl; wxl, waxyl.

www.plantphysiol.orgon May 24, 2018 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

714 Giroux et al. Plant Physiol. Vol. 106, 1994

In potato, expression of one of two starch AGI’ genes was increased in response to elevated SUC concentrations (Muller- Rober et al., 1990). With regard specifically to maize starch biosynthetic genes, Koch et al. (1992) showed that expression of wild-type alleles of shl and susl, the genes encoding the two forms of SUC synthase, was modulated in maize roots by sugar content. The amount of SUC leading to maximal shl expression was less than that for susl. These observations, taken together, suggest that the response of these genes relates to their role in tissues serving as a source or sink for carbohydrates. Accordingly, the maize endosperm might be expected to respond to high sugar concentrations by increasing the amount of transcripts present. In previous studies (Bae et al., 1990), we noted that mutation at either of the two structural genes (sh2 and bt2) for AGP resulted in increased steady-state levels of transcripts of the other nonallelic and nonmutant structural gene. Further studies were designed to focus on the coordination of expression between these two genes as detected by mutational analysis. It was discovered that this networking in expression is more global than just these two genes, reflects the severity of the genetic block, and involves the expression of genes important for at least starch and protein synthesis. The mutant effect can be mim- icked in wild-type seed by increases in the sugars presented to developing kemels. The results of those studies are the subject of this report.

MATERIALS AND METHODS

Nomenclature

Names and symbols for genes, their alleles, and their products follow the 1993 version of maize genetics nomenclature (Beavis et al., 1993). When reference is made to the gene or to its alleles, the abbreviation is italicized; for example, sh2 is used to refer to this locus. Alleles beginning with a capital letter, e.g. Sh2-, refer to the functional or dominant form of the allele, whereas recessive alleles are designated with a lowercase first letter, e.g. sh2-. A hyphen follows the allelic symbol and the designation of the particular allele, e.g. sh2-R, is given when known or relevant. Gene products are not italicized. Proteins are given in all capital letters, e.g. SH2. Transcripts, as suggested by Oliver Nelson, chairman of the Maize Nomenclature Committee, are distinguished by the nonitalicized gene symbol; the first letter is capitalized when the transcript arises from a functional wild-type allele.

Plant Culture

The sh2- and bt2- mutant alleles were described previously (Hannah and Nelson, 1975,1976). The W64A isogeneic lines were wild type or contained the mutant alleles, s h l - , sh2-, b t l - , bt2-, 02-, a e l - , dul-, or wxl- as well as a e l - du l - wxl-. The W64A isolines were grown in a greenhouse with tem- peratures of 33OC (day) and 24OC (night). Developing ears were utilized as a source of kemels for RNA, protein, and run-on transcription analysis.

RNA Isolation

RNA was isolated from kemels that had been quick-frozen in liquid NZ. The samples were ground to a fine powder with

a mortar and pestle, and RNA was isolated by the LiCl method. of McCarty (1986). Total RNA was ublized in the northern blot analysis.

RNA Gel Blots

RNA samples were denatured in 2.2 M formaltlehyde/50% formamide and fractionated on 1.5% agarose/2 2 M formal- dehyde gels using a Mops buffer (Maniatis et al., 1982). RNA gels were blotted onto nylon membrane (Hyborid N, Amer- sham) lollowing standard protocols (Maniatis et al., 1982). The transfer buffer was 1OX SSPE (1.8 M NaC1, 100 m~ NazHPO, [PH 7.71, and 10 m~ EDTA). After blotting, samples were cross-linked to the wet nylon membrane ~y 6 min of UV irradiation (3 J/C”).

Hybridizations were in 0.5 M Na2HP04 (pH 7.2), 7% SDS, and 1% BSA (Church and Gilbert, 1984). Blots were prehy- bridized for 1 h at 65 to 67OC, and hybridizaticm were for 16 to 20 h, also at 65 to 67OC. Probes were prepared by the random primer method (BRL protocols) to a specific activity of greater than 5 X 10’ cpm/rg cDNA. Blots were washed twice in 2X SSPE, 0.1% SDS and twice with 0.1X SSPE, 0.1% SDS. Each wash was for approximately 30 min at 65 to 67OC. Filters were dried and subjected to autoradiography. Signals were quantified using an Ambis radioartalytical im- aging system. Filters were reprobed with an rDNA clone, and data were adjusted for minor loading discrepancies. Data in Tables I, 11, and I11 using replicated RNA or nuclei samples from kemels of greenhouse-grown maize gave values that were within f25%.

Protein Analysis

Tissue for protein samples was ground to a fine powder with a mortar and pestle in liquid NZ and added to an equal weight:volume of buffer containing 0.1 M Tris-HC1 (pH 7.8), 10 r m DTT, and 20% glycerol. 2-Mercaptoethanol was added to 5%, and the samples were heated to 9OoC for 3 min. Protein determinations were as described by Bradford (1976) using a BSA standard. SDS-PAGE was by the method of Laemmli (1970). Proteins were transferred to mtrocellulose membranes (BA-85 0.45 pm, Schleicher & Schull) by standard protocols (Ausubel et al., 1987). Ponceau S staining provided visual confirmation of even protein loading and electroblot transfer (Ausubel et al., 1987).

Immunoblots were performed according to protocols supplied by Bio-Rad using 1 % BSA to block nonspecific antibody binding. A 1:300 dilution of SH2, BT2, or preimmune anti- serum in Tris-buffered saline with 1% BSA was used. The preparation of the SH2 and BT2 antibodies was described previously (Giroux and Hannah, 1994). Antibody binding was recognized by binding of the secondary antibody, goat anti-rabbit (Bio-Rad), linked to an alkaline phosphatase con- jugate and developed with the color reaction utilizing 5- bromo-4-chloro-3-indolylphosphate and nitroblue tetrazo- lium (BRL; Ausubel et al., 1987).

SH2 and BT2 protein levels were quantified using en- hanced chemiluminescence protocols and materials (Amer- sham). Development of the blots utilized the binding of the secondary antibody, goat anti-rabbit (Bio-Rad), linked to a



Interaction of Storage Product Genes of Maize Endosperm 715

Table I. Activity of AGP in endosperms and embryos of sh2- andbt2-mutants

Dry weights were obtained from 22-dpp kernels. The AGP ac-tivities in the mutants are relative to the standard W64A X 182E F2

endosperm and embryos._______________ __ __ __Relative AGP Activity

WTsh2-lsh2-Rbt2-B

0.09680.07600.06960.0652

Endosperm

1.000.380.190.24

Embryo

1.001.061.441.15

horseradish peroxidase activity. Several exposures of eachblot were developed, and bands corresponding to SH2 andBT2 protein subunits were quantified using a scanning den-sitometer (Molecular Dynamics, Sunnyvale, CA).

AGP Activity

The determination of AGP activity was performed as de-scribed by Hannah and Nelson (1975). Activity values pre-sented in Table I represent amounts relative to the standardin the presence of 20 mM 3-phosphoglyceric acid.

Isolation of Nuclei and Run-On Analysis

Nuclei were obtained from kernels that had been quick-frozen in liquid N2. The nuclei were isolated, and the run-ontranscription assays were performed according to the methodof Walling et al. (1986). Quantification of nuclei was by UVspectroscopy. Labeled transcripts were isolated according tothe method of Manley et al. (1983). Addition of 2 /ig/mL a-amanitin decreased incorporation below detection. Hybridi-zations were with 7.5 /tig of each plasmid on Southern blotsusing hybridization and washing conditions for the DNA andRNA blots. Blots were quantified with a Molecular DynamicsPhosphorimager. Values were standardized to the rDNAsignals.

In Vitro Kernel Culture

Blocks of W64A X 182E F2 kernels were isolated at 22 dppfrom the field-grown maize and surface sterilized accordingto the methods of Gengenbach (1977). They were placed onfilter paper in Petri plates moistened with filter-sterilized Sueor mannitol solutions. Plates were incubated for 4 d at 22°Cin the dark. Kernels were quick-frozen, RNA was isolated,and the transcript amounts were determined as describedabove.

RESULTS AND DISCUSSION

Expression of Starch Synthetic Genes in SelectedStarch Mutants

Previously we reported that the Sh2 transcript was elevatedin bt2- mutants and the Bt2 transcript was elevated in sh2-mutants (Bae et al., 1990). Since SH2 and Bt2 encode subunitsof one enzyme, AGP, and because the genes likely share a

common evolutionary origin (Bae et al., 1990; Bhave et al.,1990), we entertained the idea that the data from the mutantswere uncovering a form of end-product feedback control.Alternatively, the elevated transcripts could reflect a moreglobal effect, perhaps related to changes in amounts of ametabolite involved in starch synthesis. To determinewhether the elevation of Sh2 and Bt2 transcripts was con-fined to sh2- and bt2- mutants, we assayed transcript andprotein amounts in additional mutants that affect starchcontent (Fig. 1).

In agreement with previous data, the relative amount ofSh2 transcript is increased in bt2- mutants and the Bt2transcript is increased in sh2- mutants (Fig. 1A). That theincrease is not confined to AGP structural genes is also shownin Figure 1. The relative amount of Sh2, Bt2, and Shltranscripts present in sh2-, shl-, btl-, and bt2- mutants at 22dpp was analyzed by the quantification of northern blots.The largest increases in Bt2 transcript were in the sh2- andbtl- mutants, whereas no increase in shl- was observed.Likewise, the Sh2 transcript was increased in both the bt2-and btl- mutants, but in shl- it was relatively unaffected.There was a noticeable but small increase in the amount ofthe Shl transcript in btl- and btl- mutants. Since both theSh2 and Bt2 transcripts were increased in the btl- mutantand Btl is the structural gene for an adenylate carrier (Sulli-van et al., 1991), we conclude that the signal causing theincreased Sh2 and Bt2 transcripts is not the loss of thesetranscripts and functional AGP activity but rather is relatedto some other change in the starch biosynthetic pathway.

Additionally, we asked whether the increased transcriptcontent led to an increased amount of protein. The use ofSH2- and BT2-specific antibodies showed that SH2 protein

§

3 5 a sto tf) -Q *Q

* 4Sh2 cDNA Probe

tBt2 cDNA Probe

Sh1 cDNA ProbeSH2 Antibody

Figure 1. Interaction of starch biosynthetic mutants. A, Compilationof three RNA gel blots each with five lanes. The genotypes andcDNA probes are as listed. For each genotype, 7.5 /ig of total RNAwere loaded. B, Western blot of the same genotypes with 10 n% oftotal buffer-soluble protein from 22-dpp kernels loaded in eachlane. The western blot was probed with the SH2 antibody. The wildtype (WT) utilized was standard W64A x 182E F2 kernels. www.plantphysiol.orgon May 24, 2018 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.


71 6 Giroux et al. Plant Physiol. Vcd. 106, 1994

increased in both the btl - and bt2- mutants (Fig. lB), and the BT2 protein increased in sh2- and btl - mutants (data not shown). These data point to Sh2 and Bt2 transcript content as one (and perhaps the major) limiting step in the synthesis of the two AGP subunits and are consistent with the observed increase in enzymic activity associated with an increase in the number of functional Sh2- and Bt2- alleles (Hannah and Nelson, 1975).

The Sh2 and Bt2 transcripts barely increased in the shl- mutant. The product of shl , Suc synthase, comes early in the starch biosynthetic pathway, whereas the proteins of the sh2, bt2, and btl genes function much later. Possibly, elevation in a metabolite following cleavage of SUC, but before the synthesis of ADP-Glc, is involved in the response. However, shl- mutants differ from most sh2-, bt2-, and btl- mutants in that the latter are associated with a more severe block in the pathway, resulting in lower starch content and greater increases in soluble sugars. conceivably, then, the differences noted in Figure 1 reflect differences in the severity of the genetic block.

To address this question, we examined two sh2- mutant alleles of varying severity. Although most sh2 mutants are phenotypically indistinguishable, the mutant allele sh2-I, isolated by M. Neuffer, leads to a phenotype intermediate between wild type and the standard sh2-R phenotype (Han- nah et al., 1993). As a measurement of this, mature backcross- 3 kemels from segregating ears of the genotypes, sh2-1lsh2- llsh2-R and sh2-R/sh2-R/sh2-R were weighed. The sh2-1- containing kemels averaged O. 140 g/kemel, whereas the other class averaged 0.088 g/kemel. Furthermore, relative amounts of AGP activity in the endosperm and embryo and seed weights at 22 dpp were measured (Table I) in sh2-R and sh2-1 as well as in the phenotypically severe bt2- mutant, bt2- B . As expected, the sh2-1 mutant has higher AGP activity in the endosperm relative to the other AGP mutants. Embryo AGP activity was not reduced in these mutants and may in fact be increased relative to the standard W64A X 182E F2 kemels used.

The two mutant sh2- alleles also differ in the extent of elevation of the Bt2 transcript as well as the transcript of the endosperm-specific, starch synthetic gene Wxl (Table 11). We surveyed a series of 10 sh2- mutants and 8 bt2- mutants for Bt2 and Sh2 transcript amounts, respectively (data not shown). In agreement with the pattem seen with sh2-R and

sh2-I, phenotypically less severe mutants conditioned smaller increases in the transcript of the nonallelic gene When these data are considered together, it appears that the extent of increase in the nonmutant transcripts is related to the severity of the block in the starch biosynthetic pathway rather than the actual position of the biochemical lesion in i he pathway, a conclusion also supported by data presented in a later section.

To determine whether elevated transcripts íire in kemel components other than the endosperm, transaipts in sh2-R and sh2-1 embryos and endosperms were detennined (Table 11). Although small increases were noted in embryo transcripts, much larger increases in transcripts were noted in the endosperm. This was expected since the genes examined are endosperm specific and are involved in the production of endosperm storage compounds.

The source of the transcript increase in sh2-I, sh2-R, and bt2-B was addressed with run-on transcription assays (Table 111). Transcription rates of endosperm-specific starch biosynthetic genes were elevated and were comparable to the increases in the steady-state levels of transcripts. ‘The increased sh2-I transcript relative to standard W64A X 182E FZ kemels likely represents part of the response of starch genes to feedback control. Transcription of sh2-R was undetectable. sh2-R has an insertion in the 5‘ region of the transcribed region (J.R. Shaw, M.J. Giroux, and L.C. Hannah, unpublished data). We conclude that the elevation in the nonmutant transcripts is a consequence of increased transcription rather than an effect at the posttranscriptional level.

Transcript Increases Exist in Additional Starch Biosynthetic Mutants

To extend these analyses to other mutations affecting storage products of the maize endosperm, isogeneic lines containing a number of endosperm-specific mutatons were examined. The W64A isogeneic lines utilized were wild type or contained a single shl- , sh2-, btl- , bt2-, or 02- mutation. A W64A isoline containing the triple mutant a e l - dul- wxl- was also used. Table IV presents Shl, Sh2, Bt2, Wxl, S zein, and L zein transcript amounts at early (14 dpp), middle (22 dpp), and late (30 dpp) stages of development. Representative northem blots of 22-dpp kemels are shown in Figure 2. The middle stage (22 dpp) is recognized as the p:ak of starch

Table II. Bt2, 5 zein, Agp2, and Wxl transcript content of the same sh2- and bt2- mutants shown in Table I and embryo (€mbr)- and endosperm (Endo)-specific transcripts in selected sh2- mutants at 22 dpp

expressed relative to our standard wild type The transcripts were measured from northern blots containing 7.5 fig of total RNA at 22 dpp and

Relative Steady-State Transcript Amount ~ ~~ ~~

Transcript WT sh2-I sh2-R WT ~ h 2 - / sh2-R btZ-B -

Endo Embr Endo Embr Endo Embr

Bt2 1.00 3.97 6.91 0.43 1.00 1.00 3.96 1.86 6.55 1.28 S zein& 1.00 0.73 0.74 0.85 1.00 1.00 0.92 1.04 0.67 0.92 Agp2 1.00 0.83 1.89 1.28 1.00 1.00 1.25 1.14 1.81 1.34 wx 1.00 1.27 4.09 2.91 1.00 1.00 1.79 1.44 3.70 1.96



Interaction of Storage Product Genes of Maize Endosperm 71 7

Table 111. Transcription rates of endosperm-specific genes in selected sh2- and bt2- mutants Nuclei were extracted from intact kernels at 22 dpp. Transcription rates were expressed relative

to standard W64A X 182E Fz. Data are the averages of two or more experiments. so values ranged from 1 to 23%.

Relative Transcription Rates

WT sh2-I sh2-R bt2-B Transcript

Shl 1 .o0 0.93 1.10 1.33 Sh2 1 .o0 2.62 O 4.82 Bt2 1 .o0 7.64 7.08 O Wxl 1 .o0 3.54 3.28 6.08 S zein 1 .o0 1 .O6 0.91 0.81

biosynthesis (Ingle et al., 1965; Tsai et al., 1970). Mutant transcript content is expressed as a proportion of that in the standard W64A kemels in which transcripts of the starch synthetic genes, per total RNA, are highest at 14 dpp, remain high, and then rapidly decline at approximately 25 dpp (data not shown). Amounts at 30 dpp represent only 5 to 15% of those seen at 14 dpp.

If the data are expressed on a protein or endosperm basis, a peak at 22 dpp occurred with most transcripts. Furthermore, AGP and Sh2 and Bt2 transcript developmental profiles were virtually superimposable in wild type when the data were expressed on a kemel or protein basis (data not shown). The decline in Sh2 and Bt2 transcripts occurred more rapidly than that reported recently by Prioul et al. (1994); however, this most likely represents differences in growing conditions.

At 14 dpp, the greatest increases in the endosperm-specific transcripts were in btl-, bt2-, and a d - dul- wxl- mutants (Table IV). These mutants severely block the starch biosynthetic pathway. The Bt2 and Wxl transcripts were the most highly elevated at this stage of development. The increases relative to wild type were a maximum of 478% in the btl - mutant for the Bt2 transcript. Shl, S zein, and L zein transcripts decreased or were only moderately increased.

The increases at 14 dpp were accentuated at 22 dpp (Table IV). The greatest increases (more than 5-fold) were again observed in the mutants in which starch synthesis is severely blocked. At 30 dpp, this pattem was amplified (Table IV). Increases of 10-fold or more occurred in the sh2-, btl- , bt2-, 02-, and the ae l - dul- wxl- mutants. The Sh2 and Wxl transcripts were most affected.

A surprising observation was the increased accumulation of the starch synthetic transcripts in the 02- mutant. Increases of approximately 10-fold were noted in the Shl and Sh2 transcripts at 30 dpp. Since 0 2 is involved in zein biosynthesis with no reported effects on starch biosynthesis, this result was unexpected. Since starch synthesis transcripts were altered in the 02- mutant, we asked whether zein transcripts might be altered in the starch mutants.

At 14 dpp, there were small increases in S and L zein transcripts in sh2- (Table IV). Decreases in these transcripts relative to the standard W64A were observed in the remain- der of the mutants. At 22 dpp, there were increases in the zein transcripts in most starch mutants (Table IV). The greatest increases were in btl- and in btZ-. Greater than 10-fold increases occurred in shl- , wxl-, and ael-dul-wxl- at 30 dpp

in S and L zein transcripts (Table IV). The lowest overall increases were in btl - and 02-. Large increases in S zein transcript were not unexpected since 0 2 affects mainly the production of 22-kD (L) zein proteins.

The expression of genes not known to affect the production of storage products in the maize endosperm were also fol- lowed in the mutants. Clones of the genes, agpl and agp2, embryo counterparts of sh2 and bt2, respectively (Giroux and Hannah, 1994), and susl, which encodes Suc synthase primarily in the embryo (McCarty et al., 1986), were used to compare transcripts in 22- and 30-dpp kemels (Table IV). Although there were some increases in these transcripts at 22 dpp (e.g. a 4-fold increase in Agpl transcript in ae l - du l - wxl-), the increase in transcripts conditioned by these mutants was generally less than that observed for endosperm- specific storage product genes. However, increases in the embryo genes would be expected if they were induced by mechanisms that affect the kemel as a whole and that were not simply endosperm specific. Although the sh2 and bt2 genes are endosperm specific and do not reduce the AGP activity in the embryo, sugar concentrations are elevated in mutant embryos. TB-A translocations were used to generate seed having a wild-type embryo but a sh2- mutant endosperm. Embryos of such seed had an elevated sugar content (C. Parera, D. Cantliffe, D.R. McCarty, and L.C. Hannah, unpublished data).

SH2 and BT2 Proteins Are Not Elevated to the Same Extent as Their Transcripts in the Starch Mutants

Because of the magnitude of the increase in transcripts noted in the mutants conditioning severe blocks in starch synthesis at 30 dpp, we asked whether the cognate protein was also increased to the same extent. SH2 and BT2 proteins were measured in the isogeneic mutants at 14 and 30 dpp (Fig. 3). Although the proteins were elevated in some of the severely blocked starch mutants (b t l - and the triple mutant ae l - dul- wxl-) as well as wxl- and dul-, the increase was not nearly as great as noted for the transcripts. Mutants not increasing transcripts showed no increase in proteins. The approximate 2-fold increase in the amount of SH2 and BT2 protein in btl- is in agreement with the reported elevated AGP activity in developing btl - kemels (Doehlert and Kuo, 1990).

Interestingly, both BT2 and SH2 proteins were reduced in



71 8 Giroux et al. Plant Physiol. Vol. 106, 1994

Table IV. Relative transcript amounts in endosperm-specific and nonendosperm-specific genes in W64A isogeneic lines a t 74, 2í, and 30 dpp

was not determined. WT, Wild type. Data are the averages of two or more experiments. SD values ranged from 2 to 26%. Transcript amounts were quantified from northern blots prepared from 7.5 f ig of total RNA at 14, 22, or 30 dpp. NA, Transcript amount

Line Transcript

ae l - dul- wxl- WT shl- sh2- bt i - bt2- 02-

Shl Sh2 Bt2 W x l S zein L zein

Shl Sh2 Bt2 W x l S zein L zein AgP 1 AgP2 sus

Shl Sh2 Bt2 W x l S zein L zein AgPl AgP2

1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0

1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0

1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0 1 .o0

0.04 1.11 1.86 0.64 0.78 0.70

0.28 1 .o0 2.32 2.18 2.29 1 .O7 2.87 1.99 1 .O5

0.65 9.28 4.61 6.93

19.00 25.90

2.76 1.85

1 .O6 0.1 1 2.16 1.23 1.25 1.52

2.30 0.19 5.16 5.43 1.60 0.67 3.23 2.10 0.95

3.82 0.52 4.91

12.19 7.94 6.00 NA NA

14 dpp

0.56 1.88 4.78 2.92 0.81 0.76

22 dpp

3.43 5.40 4.27 6.56 3.06 1.36 2.95 1.81 1.22

30 dPP

5.10 10.45 3.80 4.96 4.01 5.35 2.67 1.32

Susl 1 .o0 2.02 NA 1 .O7

0.84 3.09 0.21 2.73 0.71 0.59

4.72 6.72 0.25 8.36 3.08 1.49 4.57 2.84 1.33

7.73 20.98 0.12

16.09 11.99 8.05 2.12 1.35 1 .52

0.50 1.11 1.94 1.10 0.29 0.13

2.85 2.46 2.40 1.50 0.50 0.04 3.09 1.96 1.17

9.12 10.14 2.16 5.18 9.2 1.39 3.73 1.90 1 .o9

0.62 1.74 3.54 0.1 5 0.60 0.37

2.37 4.74 5.51 0.30 1.86 0.71 4.22 2.40 1.37

9.98 33.63

9.1 1

18.53 13.20 4.34 2.15 1.91

2.115

amount in the bt2- and sh2- mutants at 30 dpp. The BT2 protein is, in fact, synthesized in sh2- and the SH2 protein is synthesized in bt2- (Fig. 3). At 14 dpp, the BT2 protein in sh2- is comparable (83%) to wild type and the SH2 protein in bt2- is comparable (117%) to wild type. Their abundance at 14 dpp but decline at 30 dpp, relative to that in wild type and the other mutants, suggests that the SH2 and BT2 proteins are more susceptible to degradation when the for- mation of a SH2/BT2 polymer does not occur.

Transcription Increases Coincide with Increased Transcripts

Nuclei isolated from standard W64A, b t l - , bt2-, and wxl- mutant kemels at 30 dpp were used in nuclear run-on assays, and transcription of the Shl, Sh2, Bt2, Wxl , and S zein genes were monitored. Figure 4 presents transcriptional activity in the mutants relative to the standard W64A. The increases are of the same magnitude as the increased steady-state transcript content. Increased transcription rates apparently account for the increase in steady-state transcript, as has been shown for hordein B and C genes (Sorenson et al., 1989).

Influence of SUC and Mannitol on Transcript Amounts

The increased transcripts in the starch-deficient mutants could be due to a buildup in a metabolite(s) in the synthesis of starch or to lower starch content. In an attempt to distin- guish between these possibilities, an in vitro kemel development procedure was used to alter the amount of SUC entering wild-type seed. If the response in the mutants is due to deficient starch content, we expect that transcripts would decrease as SUC in the growth medium increases from suboptimal to optimal concentrations. Conversely, if the mutant response is due to elevated sugar concentrations, a positive correlation between transcript amounts and sugars would be expected.

Cobb and Hannah (1983) reported that maize kemels grown in vitro develop normally, resemble field-grown kernels, and develop to maturity at 30 to 35 dpp when supplied with relatively high SUC. In subsequent studies, Cobb et al. (1988) showed that lower Suc concentrations (0.06-0.3 M) led to suboptimal growth, as judged by lowered kemel weight.

Table V shows the resulting transcript content when seeds



Interaction of Storage Product Genes of Maize Endosperm 719

Transcript2!

(DCD

Bt2 ••ff

Wx1

L zein

Figure 2. Representative northern blots of isogeneic developingkernels in a W64A background harvested at 22 dpp. These blotsare representative of the data presented in Table IV. Area shown isthe region of hybridization for each of the listed transcripts.

« I « s s i I

WT Nuclei

Transcript

Sh1Sh2Bt2WxS zein

fbt2- Nuclei

Genotype of Nuclei

WT bt1- bt2- wx1-

.00

.00,00,00

1.00

3.6010.489.806.082.74

5.274.440.169.580.54

2.727.552.374.002.03

S i * a

SH2 Antibody 14 dpp SH2 Antibody 30 dpp

Sh2 Protein Level1.34 0.64 1.17 0.00 1.00 1.78 1.43 0.35 0.32 1.00

0.95 1.48 1.10 0.61 1.51 0.82 1.79 0.64

C ...... __- D

BT2 Antibody 14 dpp BT2 Antibody 30 dpp

Bt2 Protein Level1.47 0.63 0.10 0.83 1.00 1.99 1.65 0.37 0.33 1.00

1.36 2.12 1.20 0.69 1.59 0.86 1.90 0.60

Figure 3. Western analysis of the SH2 and BT2 proteins in isogeneiclines at 14 and 30 dpp. The relative protein content of SH2 andBT2 was determined in the W64A genotypes as listed for eachfigure. Total buffer-soluble protein (20 #tg) was loaded for eachgenotype in A, B, C, and D. The SH2 antibody was used in A andB and the BT2 antibody was used in C and D. A and C representprotein samples from 14-dpp kernels and B and D represent proteinsamples from 30-dpp kernels. Data are the average of two experi-ments and are expressed relative to the standard W64A (WT). sovalues ranged from 0 to 21%.

Figure 4. Relative run-on transcription at 30 dpp in isogeneicnuclei. A and B are Southern blots with lanes containing Sh1, Sh2,Bt2, Wxl, and S zein cDNAs. Blot A was probed with labeledtranscripts from wild-type nuclei, and blot B was probed withtranscripts from 6(2— nuclei. Blots probed with RNA from bt1— andivx/— are not shown. An rDNA clone was used as a control for thetotal amount of hybridization and as a marker for correct filmexposures. The counted values from a Molecular Dynamics Phos-phorimager were corrected based on the rDNA value for each blot,and final corrected transcription rate values are presented in C.Data presented in C are averages of two or more experiments, sovalues ranged from 10 to 28%.

developed on varying concentrations of Sue or mannitol. Sh2and Bt2 transcripts increased as the Sue in the growth me-dium was elevated, whereas the Shl and S zein transcriptswere much less responsive. This differential response of thetranscripts to alterations in Sue in the growth medium mimicsthe alterations in transcripts of the high sugar/low starchmutants (sh2-, bt2-, btl-, and the triple mutant ael- dul-wxl-). Furthermore, since the low concentrations of Sue inthe growth medium give rise to reduced seed weight (Cobbet al., 1988) and presumably lowered starch content, thesignal leading to the elevation in transcripts may involve anincrease in sugars rather than the reduced starch content.The transcripts most increased, Sh2, Bt2, and Wx, were alsothose most involved in the type and amount of starch pro-duced. This argues that genes directly involved in producingstorage products in the endosperm are most sensitive to anincrease in metabolically active sugars. Sh2 and Bt2 tran-scripts were also modulated by varying mannitol in thegrowth medium. Although the pattern of change was lesspronounced than that seen with Sue, the alteration in tran-script amounts may be a response to altered osmotic condi-tions rather than a direct effect of Sue.

Sugar effects have been reported for other genes as well. www.plantphysiol.orgon May 24, 2018 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.


720 Giroux et al. Plant Physiol. Vol. 106, 1994

Table V. Relative Shl, Sh2, Bt2, and S zein transcripts in wild-type kernels grown on different concentrations of Suc or mannitol

S h l , Sh2, Bt2, and S zein transcripts were determined in W64A kernels at 22 and 26 dpp. Kernels that were 22 dpp were isolated and incubated for 4 d o n 0.06, 0.3, 0.6, and 0.9 M SUC or mannitol. The transcript contents relative to t h e levels in the 22-dpp field material were determined from RNA gel blots containing 7.5 pg of total RNA. The blots were probed with S h l , Sh2, Bt2, and S zein cDNAs. The transcripts are expressed relative to the field material at 22 dpp. SD values ranged from o to 10%.

Transcript

Shl Sh2 Bt2 S zein Treatment (hi)

S U C 0.06 0.53 0.34 0.25 0.44 0.30 0.74 0.66 0.62 0.54 0.60 0.72 0.91 1 .o1 0.37 0.90 0.42 0.91 0.77 0.46

0.06 0.66 0.52 0.37 0.47 0.30 0.91 0.81 0.80 0.87 0.60 0.66 0.52 0.39 0.65 0.90 0.55 0.34 0.35 0.52

Field 26 dpp 0.73 0.72 0.75 0.81

Mannitol

Field 22 dpp 1 .o0 1 .o0 1 .o0 1 .o0

For example, the transcriptional activity of maize photosyn- thetic promoters is repressed by SUC, Glc, and acetate in maize protoplasts relative to a mannitol control (Sheen, 1990). More germane to our system, Koch et al. (1992) showed that Shl and Susl expression in maize root tips is modulated by sugar concentration, with each gene exhibiting a distinctive profile. Taken together, the evidence indicates that the effects of sugars may be related to the identity of the tissue as a source of or a sink for carbohydrates, as has been suggested by others. Genes involved in the production of SUC would be repressed by SUC and those involved in the production of starch from SUC would be stimulated. Such an explanation may explain part of the global results presented here in the starch and storage protein mutant backgrounds.

Endosperm Storage Product Genes Are Coordinately Regulated

Coordinate gene expression is required for the successful development of any organism, including the maize kemel. Our data show that expression of many genes involved in endosperm development is sensitive to perturbations in developmental or biochemical processes seemingly not directly related to the affected gene. We show that transcript accumulation of genes involved in starch and storage protein synthesis is elevated when the synthesis of these end products is impaired. The alteration is at the transcriptional level, is dependent on the severity of the lesion, and does not alter the expression of the affected genes to the same degree. The pronounced elevation in Sh2 and Bt2 transcripts is associated with increases in the cognate proteins, albeit of a much lesser magnitude, The genetic lesion, at least for starch synthesis, can be simulated by alterations in the concentration of SUC fed to the kemel. Elevation in sugars or alteration in osmotic conditions, rather than the depletion in starch, apparently is involved in the signal transduction pathway.

The identification of the primary signal tiggeMg this response awaits further investigation. Seemingly, the primary signal is not SUC, since transcripts are altered in inutants (e.g. 02 and wx) not known to significantly alter sugir content. It is interesting to note that 02- kemels dry down more slowly than their wild-type counterparts on segregating and other- wise isogeneic developing ears (M. Lopes and B. Larkins, personal communication). Slower kemel dry-down is a well- known pleiotropic effect of the starch mutants, presumably because of the ability of the sugars to increase the moisture- holding capacity. Perhaps transcription of these storage- product genes is affected by the osmotic potential of the endosperm.

Although zein transcripts are elevated in the severely blocked starch mutants, zein protein is greatly reduced (Bar- bosa and Glover, 1978). Clearly, this reduction u1 zein is not due to a lack of zein transcripts. Furthermore, lhe increased transcripts of the starch synthetic genes in 02- may be related to the increase seen in the non-zein proteins.

Although SH2 and BT2 proteins are elevated in some mutants primarily at the latter stages of development, the extent of the increase is much less than that noted for their transcripts. This result is unexpected, since AGP activity shows a linear increase with the number of functional Sh2 or Bt2 alleles (Hannah and Nelson, 1975). Collectively, these data point to mechanisms regulating protein amount that are, or can become, independent of transcript abundance. This is similar to a case described for the Shl gene in which transcripts increase in response to anaerobiosis without a concom- itant increase in protein (Talliercio and Chourey, 1989). How- ever, it contrasts with the anaerobic induction of SUC synthase transcription and translation in rice (Ricard et a1 ., 1990).

Aside from the complexity mentioned above, our data also suggest that the rates of SH2 or BT2 protein tumover are dependent on the presence of a BT2 or SH2 protein, respectively. The SH2 protein is present in wild-type amounts in



Interaction of Storage Product Genes of Maize Endosperm 72 1

14-dpp bt2- seeds; yet in 30-dpp bt2- kemels, it is greatly reduced relative to wild type and the other mutants. The identical pattem was seen with the BT2 protein in sh2- mutants. These observations are reminiscent of an Arabidopsis mutant (Lin et al., 1988) that lacks both subunits of AGP. Although this mutation may lie in a regulatory gene involved in AGP synthesis, as the authors suggest, it seems equally plausible that the mutation is in a structural gene and the subunit of the nonallelic, wild-type gene is unstable in the absence of the other subunit.

ACKNOWLEDGMENTS

The Waxy1 cDNA probe was generously provided by Susan Wessler (University of Georgia). Quantification of blots, both with the AMBIS and the Molecular Dynamics Phosphorimager and scanning densitometer, was done in a facility of the Interdisciplinary Center for Biotechnology Research (University of Florida).

Received March 23,1994; accepted June 29, 1994. Copyright Clearance Center: 0032-0889/94/106/0713/10.

LITERATURE CITED

Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (1987) Current Protocols in Molecular Biology. Greene Publishing and Wdey-Interscience, New York

Bae JM, Giroux M, Hannah L (1990) Cloning and characterization of the brittle-2 gene of maize. Maydica 3 5 317-322

Barbosa HM, Glover DV (1978) Genes and gene interactions affecting protein and lysine content in the endosperm of maize. Braz J Genet 1: 29-39

Beavis W, Berlyn M, Burr B, Chandler V, Coe E, Nelson O (1993) A standard for maize genetics nomenclature. Maize Genet News- lett 67: 171-173

Bhave MR, Lawrence S, Barton C, Hannah LC (1990) Identification and molecular characterization of shrunken-2 cDNA clones of maize. Plant Cell 2 581-588

Boyer CD, Preiss J (1978) Multiple forms of starch branching enzymes of maize: evidence of independent genetic control. Biochem Biophys Res Commun 8 0 169-175

Bradford MM (1976) A rapid and sensitive method for the quanti- tation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 7 2 248-254

Chourey PS, Nelson OE (1976) The enzymatic deficiency conditioned by the shrunken-1 mutations in maize. Biochem Genet 1 4

Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad

Cobb BG, Hannah LC (1983) Development of wild type, shrunken- I, and shrunken-2 maize kemels grown in vitro. Theor Appl Genet

Cobb BG, Hole DJ, Smith JD, Kent MW (1988). The effects of modifying sucrose concentration on the development of maize kemels grown in vitro. Ann Bot 6 2 265-270

Creech RG (1965) Genetic control of carbohydrate synthesis in maize endosperm. Genetics 52: 1175-1186

Doehlert DC, Kuo TM (1990) Sugar metabolism in developing kemels of starch deficient endosperm mutants of maize. Plant Physiol92: 990-994

Gengenbach BG (1977) Development of maize caryopses resulting from in-vitro pollination. Planta 134 91-93

1041-1055

Sci USA 81: 1991-1995

65: 47-51

Giro- MJ, Hannah LC (1994) Expression of ADP-glucose pyrophosphorylase genes in shrunken2 and brittle2 mutants of maize. Mol Gen Genet 431: 400-408

Habben JE, Kirleis AW, Larkins BA (1993) The origin of lysine- containing proteins in opaque-2 maize endosperm. Plant Mol Biol

Hannah LC, Girow M, Boyer CD (1993) Biotechnological modifi- cation for sweet com and maize improvement. Sci Hortic 55:

Hannah LC, Nelson OE Jr (1975) Characterization of adenosine diphosphate glucose pyrophosphorylases from developing maize seeds. Plant Physiol55 297-302

Hannah LC, Nelson OE Jr (1976) Characterization of ADP- glucose pyrophosphorylase from shrunken-2 and brittle-2 mutants of maize. Biochem Genet 1 4 547-560

Ingle J, Beitz D, Hageman RH (1965) Changes in composition during development and maturation of maize seeds. Plant Physiol 40:

Koch KE, Nolte KD, Duke ER, McCarty DR, Avigne WT (1992) Sugar levels modulate differential expression of maize sucrose synthase genes. Plant Cell 4 59-69

Laemmli V (1970) Cleavage of structural proteins during the assem- bly of the bacteriophage T4. Nature 227: 680-685

Langridge P, Pintor-Toro JA, Feix G (1982) Transcriptional effects of the opaque-2 mutation of Zea mays L. Planta 156 166-170

Lin T-P, Caspar T, Somerville C, Preiss J (1988) Isolation and characterization of a starchless mutant of Arabidopsis thaliana lacking ADPglucose pyrophosphorylase activity. Plant Physiol86:

Ma Y, Nelson OE (1975) Amino acid composition and storage proteins in two new high-lysine mutants in maize. Cereal Chem

Maniatis T, Fritsch EF, Sambrook J (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Manley JL, Fire A, Samuels M, Sharp PA (1983) In vitro transcription: whole-cell extract. Methods Enzymol 101: 568-582

McCarty DR (1986) A simple method for extraction of RNA from maize tissues. Maize Genet Newslett 6 0 61

McCarty DR, Shaw JR, Hannah LC (1986) The cloning, genetic mapping, and expression of the constitutive sucrose synthase locus of maize. Proc Natl Acad Sci USA 83: 9099-9103

Mertz ET, Bates LS, Nelson OE Jr (1964) Mutant gene that changes protein composition and increases lysine content of maize endosperm. Science 145: 279-280

Muller-Rober BT, Kobmann J, Hannah LC, Willmitzer L, Sonne- wald U (1990) One of two different ADP-glucose pyrophosphorylase genes from potato responds strongly to elevated levels of sucrose. Mol Gen Genet 224 136-146

Nelson OE, Mertz ET, Bates LS (1965) Second mutant gene affecting the amino acid pattem of maize endosperm proteins. Science 150

Nelson OE, Rmes HW (1962) The enzymatic deficiency in the waxy mutant of maize. Biochem Biophys Res Commun 9 297-300

Prioul J-L, Jeannette E, Reyss A, Nicole G, Girow M, Hannah LC, Causse M (1994) Expression of ADP-glucose pyrophosphorylase in maize (Zea mays L.) grain and source leaf during grain filling. Plant PhysiollO4 179-187

Ricard 8, Rivoal J, Spiteri A, Pradet A (1990) Anaerobic stress induces the transcription and translation of sucrose synthase in rice. Plant Physiol95 669-674

Schmidt RJ, Burr FA, Aukerman MJ, Burr B (1990) Maize regulatory gene opaque-2 encodes a protein with a 'leucine-zipper" motif that binds to zein DNA. Proc Natl Acad S a USA 87: 46-50

Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell 2: 1027-1038

2 3 825-838

177-197

835-839

1131-1 135

5 2 412-418

1469-1470



722 Giroux et al. Plant Physiol. Vol 106, 1994

Sorenson MB, Cameron-Mills V, Brandt A (1989) Transcriptional and post-transcriptional regulation of gene expression in developing barley endosperm. Mo1 Gen Genet 217: 195-201

Sullivan TD, Strelow LI, Illingworth CA, Phillips RL, Nelson OE Jr (1991) The maize brittle-1 locus: molecular characterization based on DNA clones isolated using the dSpm-tagged brittle-1- mutable allele. Plant Cell 3 1337-1348

Talliercio EW, Chourey PS (1989) Post-transcriptional control of sucrose synthase expression in anaerobic seedlings of maize. Plant Physiol90 1359-1364

Tsai CY, Larkins BA, Glover DV (1978) Interactions of the opaque- 2 gene with starch-forming mutant genes on the synihesis of zein in maize endosperm. Biochem Genet 1 6 883-896

Tsai CY, Salamini F, Nelson OE (1970) Enzymes of carbohydrate metabolism in the developing endospexm of maize. I’lant Physiol

Walling L, Drews GN, Goldberg RB (1986) Transaiptional and post-transcriptional regulation of soybean seed protein mRNA levels. Proc Natl Acad Sd USA 8 3 2123-2127

4 6 299-306



Documents

Coordinated Transcriptional Regulation of Storage Product ... · PDF fileCoordinated Transcriptional Regulation of Storage Product ... subject of this report. ... formamide and fractionated